3 14

M. I. Kamboh ef. al.

Eleclrophoresis 1990,11,3 14-3 18

181 Wrig1ey.C. W., Autran, J . C. andBushuk,’W.,in: Pomeranz,Y.(Ed.), Advances in Cereal Science and Technology, AACC, St. Paul, MN 1982, pp. 211-259. 191 Khan, K., Bakers Digest 1982,56, 14-19.

Mohammad I. Kamboh Lori J. Kelly Robert E. Ferrell Department of Human Genetics Graduate School of Public Health University of Pittsburgh, Pittsburgh, PA

1 101 Lafiandra, D. and Kasarda,D.D., Cereal Chem. 1985,62,3 14-3 19. I 1 11 Courvoisier, C., Th&e d’Etat, Universited’Orsay, Orsay, June 1984, NO. 2914, pp. 7-135.

Genetic studies of human apolipoproteins: XIV. A simple agarose isoelectric focusing gel method for apolipoprotein E phenotyping A new and simplified procedure is described for apolipoprotein E ( A P 0 E) phenotyping from native plasma or serum samples. Diluted or dialyzed samples are separated on agarose isoelectric focusing gels followed by protein blotting on nitrocellulose membranes. APO E banding patterns are localized immunologically using polyclonal goat anti-APO E antiserum as the primary antibody and rabbit anti-goat IgG conjugated with alkaline phosphatase as the secondary antibody. The method was used in parallel with our previously described polyacrylamide gel system to screen 110 unrelated and healthy US whites. Both gel systems gave identical APO E phenotypes, and allele frequencies were comparable with reported US white values. This simplified method can be used on a large number of population and clinical samples with minimum cost and effort.

1 Introduction Apolipoprotein E (APO E) is a plasma protein that plays a pivotal role in lipid metabolism as a slructural component of various lipoproteins, and as a ligancl for two lipoprotein receptors which mediate the transport oflipid particles across the cell membrane f 11. The structural locus for APO E has been mapped to chromosome 19q13.2-(113.3 121 and is polymorphic with three common and several rare alleles [3-61. Among the three common alleles the APO E*2 and APO E*4 are functionally different from the commonAPOE*3 allele IS, 7-81. The APO E*2 allele product binds defectively to the low density lipoprotein receptor which resullts in reduced in vivo catabolism, whereas the APO E*4 allele product shows an increased in vivo catabolism. The role of APO E genetic variation in determining interindividual differences in total serum cholesterol and low density lipoprotein (LDL)-cholesterol levels in the general population is well documented [5,91. The APO E*2 and APO E*4 alleles are associated with lower and higher cholesterol levels, respectively, in populations at large. Conventionally, APO E genetic polymcrphism has been determined by isolation of apo very low density lipoprotein (VLDL) by prolonged ultracentrifugation and delipidation followed by isoelectric focusing (IEF) or two-dimensional electrophoresis and protein staining. Due to their tedious nature these conventional methods are not suitable nor economical for general population and epidemiological studies. Recently, Correspondence: Dr. M. I. Karnboh, Department of Human Genetics, GraduateSchool ofpublic Health, University ofpittsburgh, Pittsburgh, PA 15261, USA Abbreviations: APO, apolipoprotein; IEF, isoelectric focusing; TBS,Trisbuffered saline 0VCH Verlagsgesellschaft mbH, D-6940 Weinheim, IY90

three groups have reported APO E phenotyping on IEF-immunoblot gels by eliminating the ultracentrifugation step 110- 121. However, these methods still require serum delipidation [ 101, or serum treatment with guanidine-HCL 1 1 11 to dissociate APO E from lipoproteins, or charge-shift electrophoresis of hydrophobic serum proteins 121 followed by prolonged IEF. Recently, we 131 have described a simple and rapid IEF-immunoblotting method using polyacrylamide as a supporting medium to screen APO E polymorphism directly from dialyzed serum or plasma samples without prior ultracentrifugation and delipidation. In our previous method [I31 dialysis of samples was found to be necessary to obtain a clear background on polyacrylamide I E F gels. In this paper we provide evidence that APO E phenotypes can be detected easily and efficiently on an IEF agarose gel matrix using either dialyzed or diluted samples and it can be used as a complimentary method to our previously described polyacrylamide IEF method.

2 Materials and methods 2.1 Blood samples Blood samples from unrelated and healthy US white blood donors were drawn at the Central Blood Bank of Pittsburgh in 1984. EDTA plasma was then removed and stored at -70 O C . These plasma samples have been thawed and refrozen several times for typing of other genetic markers. The test plasma samples were diluted I :8 in deionized water or dialyzed against phosphate buffer, pH 6.8 (0.038 M NaH,PO,.H,O, 0.029 M Na,HP04, 0.004 M tetrasodium EDTA), using a continuous flow microdialysis system (Model 1200, Bethesda Research Laboratories) [ 131. 0173-0835/90/0404-0314$2.50/0

Elecrrophoresis 1990, I / . 314-318

2.2 Agarose gels and IEF Thin-layer agarose gels were prepared by dissolving 0.25 g Agarose IEF (Pharmacia), 3.0 g sorbital in 23 mL deionized water. The mixture was boiled and stirred continuously until a clear solution was achieved. Pharmalyte carrier ampholytes, pH 4.5-5.4, (0.375 mL) and pH 5-8 (0.75 mL), were mixed with the hot solution. The mixture was poured immediateiy between two glass plates separated by a 0.5 mm gasket. To ensure an even distribution of the agarose solution the bottom glass plate with the gasket was placed on a leveling table. The gel was left at room temperature for 30-45 min or overnight, then the upper glass plate upon which the gelation had occurred was lifted with the help of a scalpel. The surface of the gel was blotted for 5 s with a piece of Whatman 1 filter paper to remove the excessive moisture from the gel. However, this step was not always necessary. Electrolyte strips saturated with 1 M sodium hydroxide for thecathode and 1Mphosphoric acid for the anode were placed along the length of the gel. The gel was placed in an LKB model 22 17 Ultrophor IEF unit at 10 "C and prefocused for 15 rnin at 500 V, 10 W and 50 mA. Diluted and/or dialyzed samples were applied 0.5 cmfrom the cathode using 5 x 7 mm Whatman 3 MM paper wicks. IEF was continued at I000 V, 15 W, 50 mA for 30 min; sample wicks were removed and the voltage was increased to 1500 V for 90 min.

2.3 Immunoblotting After the completion of IEF, the gel was rinsed briefly with Tris-buffered saline (TBS) (0.25 M NaCI, 0.03 M Tris-HC1, pH 8.0). A 9 x 23.5 cm nitrocellulose filter (0.20 pm, Schleicher and Schuell, or 0.22 ,urn, Micron Separations Inc.)

APO phenotyping with an agarose isoelectric focusing method

3 15

presoaked in TBS was laid on the gel surface followed by a 9 x 23.5 cm Whatman 3 MM filter paper also presoaked inTBS, a stack of paper towels, a glass plate, and about 2 kg weight. The nitrocellulose filter was removed after 90 min, washed briefly in TBS, and incubated with 2-5 o/o w/v nonfat dry milk dissolved in deionized water for 60 min. The filter was then exposed for 90- 120 rnin to goat anti-human APO E polyclonal antiserum (Daiichi Chemical) diluted 1:750 in TBS buffer followed by three 10 rnin washes in TBS. The filter was then probed for 90-120 rnin to rabbit anti-goat conjugated with alkaline phosphatase at 1 5000 dilution followed by extensive washes in TBS. Finally the filter was stained histochemically using 25 mg P-naphthyl phosphate (Sigma), 25 mg Fast Blue BB salt (Sigma) and 60 mg magnesium sulfate in 50 mL stock buffer (1.8 g NaOH, 3.7 g boric acid/L).

3 Results In our early experiments we tried various single range carrier ampholytes for APO E typing on agarose IEF gels using dialyzed and/or diluted samples. The Pharmalyte pH 4-6.5 range was found to be the best single range choice as it resolved clearly all the six common APO E phenotypes using diluted, dialyzed or neuraminidase-treated samples (Fig. 1). However, since the focusing position of the E4 gene product is at the extreme basic side on the pH 4-6.5 range gel, the most cathodal band of the E4 type either migrated out of the gel or focused on the sample application area. This problem was overcome by mixing Pharmalytes 4.5-5.4 and 5-8 in a 1:2 ratio, which shifted the APO E immunoreactive bands to the middle of the gel using either diluted (Fig. 2) or dialyzed samples (Fig. 3). In the IEF gel each homozygous APO E

Figure 1. IEF-immunoblotting pattern of APD E phenotypes on agarose gels, pH 4-6.5, using dialyzed (A), diluted (B), and neuraminidase-treated (C) samples. No difference was noted on APO E banding patterns using either native or neuraminidase-treated samples.

3 16

M. I. Kamboh et. al.

Electrophoresis 1990,II,314-318

Figure 2. IEF-immunoblotting patterns of APO E phenotypes on agarose gels, p u 4.5-5.4 and 5-8, using diluted samples.

Figure 3. IEF-immunoblotting patterns of APO E phenotypes on agarose gels, pH 4.5-5.4 and 5-8, using dialyzed samples.

APO phenotyping with an agarose isoelectric focusing method

Eleclrophoresis 1990, 11, 314-318

3 17

This method is a more simplified version of our recently described rapid micro method [131. Previously we have demonstrated that A P O E phenotypes can be detected directly from plasma or serum on polyacrylamide I E F gels containing 3 M urea or sucrose as additives. However, we found that dialysis of samples was necessary in order to obtain a clear background on polyacrylamide I E F gels. Recently, we have concentrated on further simplification of the A P O E typing protocol to eliminate the need for gel additives and dialysis. However, using a continuous flow microdialysis system the dialysis step does not seem to be tedious because up to 56 Using agarose gels of this p H (pH 4.5-5.4, pH 5-8) we samples can easily be dialyzed in one set-up. We have prescreened 110 plasma samples from unrelated healthy US sented the evidence here that no prior ultracentrifugation, whites for the APO E polymorphism (Table 1). Of the six delipation or even dialysis is necessary when using the agarose possible phenotypes, resulting from three alleles, five were IEF-immunoblotting protocol. Simple dilution of samples observed in this small sample, the APO E 2-2 phenotype being with deionized water is sufficient to detect APO E banding absent. The observed values showed no significant depar- patterns on agarose I E F gels. Previously we have tried the ture from those expected. based on the Hardy-Weinberg same approach on polyacrylamide I E F gels, but due to high equilibrium &22 = 0.86 P > 0.25). Also the frequencies of the background, APO E phenotypes could not be typed. This A P O E*2, A P O E*3 and APO E*4 are similar to those background difference on agarose and polyacrylamide gels reported for other US white populations. The reliability of could be due to their different gel matrix properties. We have observed comparable APO E phenotype patterns using both APO E phenotypes obtained on agarose gels was tested by parallel running some samples on our previously described the diluted and dialyzed samples (Figs. 1-3). However, diaand well characterized polyacrylamide gel system (Fig. 4). lyzed samples tend to give a slightly clearer background than Both gel systems gave identical APO E phenotype counts. In the diluted samples. Although we have not encountered any contrast to agarose gel where APO E sialylated bands are not mistyping problems in the parallel use of dialyzed and diluted distinguishable (Figs. 1-3), the polyacrylamide gel matrix samples, one may use the dialyzed samples where the diluted clearly resolves these sialylated bands which are located more samples give high background. APO E typing can be peranodally than the asialo major isoforms (Fig. 4). Our failureto formed correctly on the agarose I E F method presented here. see sialylated bands in agarose gel is not clear; but it could be However, this method can be supplemented with our previousdue to a slightly high background in the anodal region of the ly described method on polyacrylamide gels I131 ifit is needed, especially to resolve the type 2 banding patterns which are agarose gel as compared to the polyacrylamide gel. clearly distinguishable on polyacrylamide gels (Fig. 4).

phenotype appeared as having two major bands and each heterozygote phenotype is characterized by the presence of four major bands. In fresh samples, however, a third middle band was also visualized occasionally in homozygote phenotypes. Sample dilution of 1:8 was found to be appropriate for the APO E 3-3,4-3,4-4 and 4-2 phenotypes. But atthis dilution some APO E 2-2 and 3-2 phenotypes gave high backgrounds, probably due to the high concentration of APO E associated with these phenotypes. Therefore, a high dilution, i.e. 1:10 or 1 :15, was used for these selected few samples.

4 Discussion We present a simple and specific agarose IEF-immunoblot method to screen APO E polymorphism directly from plasma or serum without prior ultracentrifugation and delipidation.

With the agarose I E F method we have tested 110 unrelated US whites for their APO E typing using both the dialyzed and diluted samples. The correct assignment of different APO E phenotypes was tested by running samples of known APO E type on each gel. Theremarkable similarity ofdifferent APO E

Table 1. Distribution of APO E phenotype and allele frequencies in US whites Phenotype

Allele frequency ~

Number tested

110

3-3 Observed Expected

4-3

3-2

68 26 13 (69.52) (24.66) (11.19)

4-4

4-2

2 (2.19)

I (1.99)

~

APOE*2 APOE*3 A P O B 4

0.064

0.795

0.141

Figure 4 . IEF-immunoblotting patterns of APO E phenotypes on polyacrylamide gels, pH 4.5-5.4 and 5-8, using dialyzed samples.

3 18

M. Kane et. al.

Electrophoresis 199O,lI, 318-321

allele frequencies between our sample and other US white populations [5-6, 131 provides further support in favor of the reliability and specificity of this method. The plasma samples we have analyzed in this investigation were collected in 1984 and have been frozen andthawedrepeatedly over this five year period for other genetic marker studies. We have used only 5 pL plasma samples for the dilution purpose. However, even a 2 pL sample is enough for a single I E F run. This method is suitable for large-scale population and epidemiological studies because only a small amount of sample is needed and no sophisticated laboratory equipment is required. To the best of our knowledge the method presented here is the first to use agarose rather than polyacrylamide as a supporting medium to detect APO E phenotypic variation. EIecause of the simple protocol presented here, IEF in agarose may prove to be the method of choice for APO E phenotyping. The present method together with our previously described procedure [ 131 provides a simple screening tool for investigation of the APO E polymorphism on a worldwide basis and the validity ofthis argument has been confirmed in our laboratory by screening several thousand plasma samples (unpublished data).

5 References [ 11 Mahley, R. W., Science 1988,240, 622-630. 121 Myklebost, 0. and Rogne, S., Hum. Genet. 1988, 78,244-247. I31 Zannis,V.l.,Just,P.W.andBreslow,J.L.,Am.J.HuniGenet.1981, 33, 11-24. 141 Wardell. M . R., Brennan, S. O., Janus, E. D., Fraser, R. and Carrell. R . W.. J . Clin. Invest. 1987. 80. 483-490. 151 Davignon, J., Gregg, R. E. and Sing, C. F.,Arteriosclerosis 1988.8. 1-21. I6 I Kamboh, M. I., Sepehrnia, B. and Ferrell, R. E., Dis.Markers 1989,7, 49-55. 171 Mahley, R. W. and Innerarity, T. L., Biochirn. Biophys. Aria. 1983, 737, 197-222. 181 Gregg, R. E., Zech, L. A., Schaefer, E. J., Stark, D., Wilson, D. and Brewer, H. B., Jr., J. Clin. Invest. 1986, 78, 815-821. 191 Utermann, G.,Arn. Heart J. 1987,113,433-440. [ 101 Menzel,H.J.andUtermann,G., Electrophoresis 1986,7,492-495. II 11 Havekes, L. M., deKnijff, P., Beisiegel, U., Havinga, J., Smith, M. and Klasen, E., J. LipidRes. 1987,28,455-463. [ 121 Steinmetz, A., J. LipidRes. 1987,28, 1364-1370. [ 131 Kamboh, M. I., Ferrell, R. E. and Kottke, B., J . Lipid Res. 1988,29, 1535-1 543. 1

This work was supported in part by N.I. H . Grants HL39107 and HL24489. We thank Jeanette Norbut for secretarial assistance. Received October 4, 1989

Masateru Kane '. Yoshio Yamamoto2 Mitsuko Yamada2 Tatsushige Fukunaga2 Yoahitsugu Tatsuno2

Phenotyping of erythrocyte acid phosphatase and esterase D by high field strength isoelectric focusing on cellulose acetate membrane

'Forensic Science Laboratory, Shiga Prefectural Police Headquaters,Ohtsu 'Department of Legal Medicine, Shiga University of Medical Science, Ohtsu

Phenotyping of erythrocyte acid phosphatase (EAP) and esterase D (ESD) by cellulose acetate membrane isoelectric focusing (CAM-IEF) under a nonequilibrium condition is described. In an attempt to improve the method of CAM-IEF, we shortened the electrode distance to provide a higher field strength at a given (low) voltage. Various carrier ampholytes for EAP typing and various chemical separators for ESD typing were also tested. Good separations were obtained after 30 rnin IEF for EAP typing and 25 min for ESD typing. When applied to blood stains and stored for various periods at room temperature, the stains up to 8 months old could still be phenotyped for EAP and those up to 4 weeks old for ESD. CAM-IEF is suitable for routine forensic work of EAP and ESD phenotyping.

1 Introduction

phenotypes are usually discriminated by either conventional electrophoresis or isoelectric focusing (IEF), broader bands In 1963 Hopkinson et al. I 1 1 first described genetic polymor- resulting from diffusion during isozyme visualization ocphism of erythrocyte acid phosphatase (EAP). Although six casionally causes mistyping. Destro-Bisol and Ranalletta 121 showed a sharper isozyme pattern by using hydrophilic Correspondence: Dr. Masateru Kane. Department of Legal Medicine, cellophane film soaked in prewarmed substrate solution at Shiga University of Medical Science, Seta-Tsukinowa-cho, Ohtsu 520-2 1. 50 OC to shorten the incubation time. Esterase D (ESD) was Japan first phenotyped by Hopkinson et al. [3] in 1973. Two rare Abbreviations: BES, N,N-bis(2-hydroxyethy1)-2-aminoethanesulfonic variants, ESD*5 and ESD*7, are not separated from three acid; CAM, cellulose acetate membrane; DTr, dithiothreitol; EAP, common phenotypes by conventional electrophoresis but are erythrocyte acid phosphatase; EPPS,N-(2-hydroxyethyl)-piperazine-N'- well separated by I E F [4,5], although the discrimination ofthe three common phenotypes by I E F is unsatisfactory. Yuasa et 3-propanesulfonic acid; ESD, esterase D; IEF, isoelectric focusing; Vh, volt x hour al. [ 61 reported that low voltage I E F under a nonequilibrium 0VCH Verlagsgesellschaft mbH. D-6940 Weinheim, I990

0173-0835/90/0404-03 18$2.50/0